In this K01 application, Kathleen Hayden PhD is seeking training in genetics with a goal of exploring cognitive endophenotypes (latent or sub-clinical phenotypes representing the underlying disease genotype) of Alzheimer disease (AD) to inform ongoing and future genetic studies. The project builds on the candidate's prior work and is focused on providing advanced expertise in the area of genetics in order to facilitate her development as an independent researcher. The plan leverages the research environment and ample resources of the Bryan Alzheimer's Disease Research Center (Bryan ADRC) and its partners, the Duke Institute for Genome Sciences and Policy (IGSP), the Center for the Study of Aging, and the Department of Biostatistics and Bioinformatics. The research plan in this application is focused on advancing methodology to enhance the detection of yet to be determined risk genes involved in AD. This application proposes to use samples from four large NIH funded studies which have all been harmonized in their clinical data collection methodologies (Cache County Memory Study, Bryan ADRC cohorts, NAS-NRC Twin Study, and the Neurocognitive Outcomes of Depression in the Elderly Study) to rigorously determine AD endophenotypes using hypothesis driven and empirically based statistical approaches. Guided by Mentors, Kathleen Welsh-Bohmer (Bryan ADRC & CCMS PI) and David Goldstein (Director of Population Genomics & Pharmacogenetics IGSP) the candidate's career development plan will also benefit from the input of collaborators Carl Pieper (biostatistics), Brenda Plassman (twin studies), and David Steffens (geriatric depression) as she strives to understand and develop latent models within the unique cohorts: The training program includes coursework in latent class analysis, latent class trajectories, genetics, and computational biology as well as experiential training in neuropsychological methods, genetic study design, and statistical approaches unique to genetic methodologies. This relevant and timely application addresses a need to apply endophenotyping methods to population genetic work. Ultimately the application of endophenotypes of AD will facilitate the identification of new genes, highlight potential causal pathways of disease, and suggest avenues for future treatment development. [unreadable] [unreadable] [unreadable]